[R] pca

[Prof Brian Ripley]

On Sun, 14 Dec 2003, Douglas Trainor wrote:

> Somewhere along the line, you have been confused.
> You're in good company though.  Factor analysis 
> and PCA are different entities entirely.

Not in SPSS, where the same command is used for both (although we were not 
told anything like enough about what was done).

*However* I don't know what is meant by `pca in R'. Standard R contains
princomp() and prcomp() to do PCA: it appears that pca() is in the
orphaned package multiv. Why not use the standard functions?  Also, those
non-default arguments would appear to be very unusual indeed for PCA, and
do not correspond to a *correlation* matrix.

> Any clues about this diference?

User error looks likely.  Please seek out local statistical expertise.

> >Dear listmates, i've done a pca analisys in R (1.8 v.) 
> >with the command
> >
> >pca(Matrix, cent=FALSE, scle=FALSE)  
> >
> >I have obtained a v matrix very different from the  
> >component matrix resulted by a factor analysis in SPSS, 
> >unrotated and with a extraction from a correlation 
> >matrix. Any clues about this diference?

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595